Saturday, September 17, 2005
gnustep on gentoo fine again
Thanks to the fixes Adam made with libgif and linubgif, core GNUstep compiles fine again on gentoo. I had to recompile/relink all applications (I don't thinkn this is nice, if the dependency is brought in by a library). Enrico also fixed also GWorkspace to compile without DBKit.. so finally the whole environment of base applications is in good shape again.
Thursday, September 01, 2005
feeling like a bug...
(the post was mysteriously lost, so I try to write it again... and of course it didn't came out the same... some of the initial feeling was lost as I thought about it again)
I watched a video by Alan Kay [ part 1 ] part 2 ] and it made me feel small and bad. And it was not because of some nostalgia that sometimes creeps up my back when I see stuff from the pioneering time of computer science. I was impressed what was done in those times! And some things can't be still done today such easily
Maybe the most impressive thing were the small programs written by children! They were nice and stimulating programs and I would have a hard time doing something alike with the tools I know (even using languages as Objective-C or Java), really! And those kids wrote one or two pages of code for drawing programs with interactive menus and design constraints? I think that controlling the computer and being able to program it is very important and today, even after almost 20 years of programming, I feel it distant and unnatural.
Another interesting thing were the interfaces. Today due to use of the Mac and the interfaces derived from IBM's CUA (motif, windows, os/2) and theyr blind clones like KDE or GNOME limit our perception. OpenView would be regarded strange (not to speak of Amapi's natural interface).
The interfaces of the programs presented tehre are instead very clean and present concepts that aren't used much today. The Alto programs striked for example with the non-intrusiveness of their menus, the modeless use... but also older programs were interesting in their use. The direct manipulation of objects in the Rand program reminds me a lot of the Newton.
We have now very powerful hardware and operating systems, but their programs run at reaosnable speed on their systems... while today often we think about excess in functions or eye-candy... and less about usability. Also thing pile up. The interface to write this blog isn't as nearly as intuitive as the text editor on the Alto was (well... it also lost my first blog entry). I write inside a browser which handles dynamic contents... the browser runs on the OS itself. The result is miserable, even if a lot of power is "stimulated" little power is "unleashed".
One last thought is about the use of the computer in education. I feel warm and cozy that Alan thinks like me: an important liberal art is important. (I translate this into: teaching to use Office in elementary school is horrible). Also kids could be teached a much more creative use of the computer as the Alto examples show. When I was 14 years old I was tought LOGO, now people at high school learn visual basic... I think I was better off! Much better!
I watched a video by Alan Kay [ part 1 ] part 2 ] and it made me feel small and bad. And it was not because of some nostalgia that sometimes creeps up my back when I see stuff from the pioneering time of computer science. I was impressed what was done in those times! And some things can't be still done today such easily
Maybe the most impressive thing were the small programs written by children! They were nice and stimulating programs and I would have a hard time doing something alike with the tools I know (even using languages as Objective-C or Java), really! And those kids wrote one or two pages of code for drawing programs with interactive menus and design constraints? I think that controlling the computer and being able to program it is very important and today, even after almost 20 years of programming, I feel it distant and unnatural.
Another interesting thing were the interfaces. Today due to use of the Mac and the interfaces derived from IBM's CUA (motif, windows, os/2) and theyr blind clones like KDE or GNOME limit our perception. OpenView would be regarded strange (not to speak of Amapi's natural interface).
The interfaces of the programs presented tehre are instead very clean and present concepts that aren't used much today. The Alto programs striked for example with the non-intrusiveness of their menus, the modeless use... but also older programs were interesting in their use. The direct manipulation of objects in the Rand program reminds me a lot of the Newton.
We have now very powerful hardware and operating systems, but their programs run at reaosnable speed on their systems... while today often we think about excess in functions or eye-candy... and less about usability. Also thing pile up. The interface to write this blog isn't as nearly as intuitive as the text editor on the Alto was (well... it also lost my first blog entry). I write inside a browser which handles dynamic contents... the browser runs on the OS itself. The result is miserable, even if a lot of power is "stimulated" little power is "unleashed".
One last thought is about the use of the computer in education. I feel warm and cozy that Alan thinks like me: an important liberal art is important. (I translate this into: teaching to use Office in elementary school is horrible). Also kids could be teached a much more creative use of the computer as the Alto examples show. When I was 14 years old I was tought LOGO, now people at high school learn visual basic... I think I was better off! Much better!
Monday, August 29, 2005
gap comes along...
Thanks to the keen help of Robert and Gregory I was able to fix so,e s,all but annoying bugs in both FTP and Graphos and thus both applications made a further step towards usability. I think that FTP is almost ready for a first "beta" release, I am waiting for some feedback, especially bug reports.
Saturday, August 27, 2005
Kaffe on Solaris/Sparc
Good news! I compiled kaffe on sparc solaris 2.6 using JIT and pthread on my dual-processor box... and yay all regression passed! included te 4 jni tests. I have never seen this on that box since years. Cheers!
Thursday, August 25, 2005
Kaffe on Darwin
Kaffe on Darwin is fine again, using interpreter and pthreads. It passes all regressions happily.
last night efforts... wx and DOM
I continued my effort to shape wxMotif in a better form on IRIX but until now no real progress... The code is ugly and although Vadim (vadz) was kind to me and spotted various methods to improve... the output was still none (ok, apparently I didn't break anything either).
The other interesting discussion I had was on IRC with Stefan. The problem is again the lack of applications in some areas and the quality of others. Today's menu was "browser". Apart from the usual talk of porting other engines... the idea of writing our own rised up again. ANd this one it seems promising.
We should start the browser work in making a sort of webcore. That one should be based on a DOM renderer, so that the translation of XML, HTML or whatever into o a DOM can be a separate module. Help in this regard could come from the Iconara DOM framework and the expat library. Makign a simle XML parser should be easy with these tools and so we could concentrate on other issues like the rendering without loosing time in the parser. Once the foundations are set, an HTML->DOM plugin could be done and added inserted into Iconara. Also this approach would force us into a good separation of tasks and thus in the future even iconara itself migh be replaced.
If this talk will have a future I don't know, but I'll think of a project code-name!
The other interesting discussion I had was on IRC with Stefan. The problem is again the lack of applications in some areas and the quality of others. Today's menu was "browser". Apart from the usual talk of porting other engines... the idea of writing our own rised up again. ANd this one it seems promising.
We should start the browser work in making a sort of webcore. That one should be based on a DOM renderer, so that the translation of XML, HTML or whatever into o a DOM can be a separate module. Help in this regard could come from the Iconara DOM framework and the expat library. Makign a simle XML parser should be easy with these tools and so we could concentrate on other issues like the rendering without loosing time in the parser. Once the foundations are set, an HTML->DOM plugin could be done and added inserted into Iconara. Also this approach would force us into a good separation of tasks and thus in the future even iconara itself migh be replaced.
If this talk will have a future I don't know, but I'll think of a project code-name!
Wednesday, August 24, 2005
The beginning
Everything has a beginning... and so also my blog. After different people showed interest in a potential blog by my side, concerning mainly my open source activities, I decided to create one and see how well I come along with it.
I do not promise regular updates... nor is the life of this blog guaranteed.
I assume that the main focus will be on my development activities in Kaffe, GNUstep... and possibly generic comments about my visions of the world of computing.
Maybe also comment on my current music, photography and vision of the world will find a way here... we shall see.
I do not promise regular updates... nor is the life of this blog guaranteed.
I assume that the main focus will be on my development activities in Kaffe, GNUstep... and possibly generic comments about my visions of the world of computing.
Maybe also comment on my current music, photography and vision of the world will find a way here... we shall see.
Subscribe to:
Posts (Atom)